|Publication number||US7499354 B2|
|Application number||US 11/268,974|
|Publication date||Mar 3, 2009|
|Filing date||Nov 8, 2005|
|Priority date||Nov 8, 2005|
|Also published as||US20070121390|
|Publication number||11268974, 268974, US 7499354 B2, US 7499354B2, US-B2-7499354, US7499354 B2, US7499354B2|
|Inventors||Theodore W. Houston, Xiaowei Deng, Tito Gelsomini|
|Original Assignee||Texas Instruments Incorporated|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (17), Classifications (12), Legal Events (2)|
|External Links: USPTO, USPTO Assignment, Espacenet|
The present invention is directed, in general, to a method for testing transistors and, more specifically, to a method for testing transistors having an active region that is common with other transistors and a testing circuit for accomplishing the same.
Modern electronic equipment, such as televisions, telephones, radios, and computers are generally constructed of solid-state devices. Solid-state devices include transistors, capacitors, resistors, and other semiconductor devices. Typically, such devices are fabricated on a substrate and interconnected to form memory arrays, logic structures, timers, and other components of an integrated circuit. One type of memory array is a static random access memory (SRAM) device, in which memory cells are continuously available for reading and writing data. As technology improves, SRAM cells and other components are fabricated at smaller sizes and with greater on-chip integration.
The increasing level of on-chip integration has allowed steady improvements in modern microprocessor performance, but has also resulted in high energy dissipation in integrated circuits. In complementary metal-oxide-semiconductor (CMOS) circuits, which are often included within SRAMs, transistor threshold voltages have been reduced, along with supply voltages, as the technology is scaled. However, decreasing the transistor threshold voltage typically increases the amount of “static” or “leakage” power dissipated by the CMOS circuit. As transistor threshold voltages continue to be reduced in emerging technologies, leakage power is becoming a sizable percentage of the total power consumed in CMOS circuits.
Knowing that the CMOS circuits may include various disadvantageous characteristics, such as the aforementioned leakage power, the industry generally desires to characterize or quantify the circuit properties before going into large scale production thereof. Such is often the case with SRAM devices. Interestingly, the characteristics of the transistors in the SRAM devices tend to be dependent on the exact environment they will ultimately be included within. For example, a transistor having a given width and length would not exhibit the same characteristics if it were isolated by itself as compared to if it were located within a high density array. Correspondingly, the industry desires to characterize the transistors of the SRAM devices in an environment similar to the high-density array environment it will ultimately be included within.
Accordingly, what is needed in the art is a method for accurately characterizing the transistors of a transistor array while the transistors are in their standard environment, and a testing circuit for accomplishing the same.
To address the above-discussed deficiencies of the prior art, the present invention provides a method for testing an electrical property of one or more functionally separate transistors located within an active region that is common with other transistors, a method for characterizing the leakage current of at least one of a plurality of functionally separate transistors located in a common active region of a circuit, and a test structure for testing one or more functionally separate transistors located within a common active region. The method for testing the electrical property, among other steps, includes providing a pair of functionally separate transistors located within an active region, and biasing a terminal between the pair relative to gates of the pair and terminals outlying the pair to obtain a leakage current associated with the pair. In one embodiment, the functionally separate transistors are configured in the same topology as in a circuit application, such as in an SRAM array.
As indicated above, the present invention further provides a method for characterizing the leakage current of at least one of a plurality of functionally separate transistors located in a common active region of a circuit. One embodiment of this inventive method includes providing a test structure comprising a replication of a plurality of functionally separate transistors located in a common active region of a circuit, the test structure thereby including a plurality of test transistors located in a common active test region, and biasing at least one pair of test transistors within the plurality of test transistors such that a leakage current of the at least one pair can be measured without contamination of spurious gate-edge-diode-leakage (GEDL) from remaining ones of the plurality of test transistors, the measured leakage current helping to characterize a leakage current of the plurality of transistors located in the common active region of the circuit.
As briefly mentioned above, the present invention additionally provides a test structure for testing one or more functionally separate transistors located within a common active region. The test structure, without limitation, includes a pair of functionally separate transistors located within a common active region. The test structure further includes a first interconnect configured to contact a terminal between the pair, a second interconnect configured to directly couple and contact gates of the pair, and a third interconnect configured to directly couple and contact terminals outlying the pair, the first, second and third interconnects configured to allow the terminal to be biased relative to the gates and outlying terminals, where the bias may be non-zero, to obtain a leakage current associated with the pair.
The foregoing has outlined preferred and alternative features of the present invention so that those skilled in the art may better understand the detailed description of the invention that follows. Additional features of the invention will be described hereinafter that form the subject of the claims of the invention. Those skilled in the art should appreciate that they can readily use the disclosed conception and specific embodiment as a basis for designing or modifying other structures for carrying out the same purposes of the present invention. Those skilled in the art should also realize that such equivalent constructions do not depart from the spirit and scope of the invention.
The invention is best understood from the following detailed description when read with the accompanying FIGUREs. It is emphasized that in accordance with the standard practice in the semiconductor industry, various features are not drawn to scale. In fact, the dimensions of the various features may be arbitrarily increased or reduced for clarity of discussion. Reference is now made to the following descriptions taken in conjunction with the accompanying drawings, in which:
The present invention is based, at least in part, on the recognition that testing the electrical characteristics of a single functionally separate transistor located within a high density array of a circuit is often problematic. The present invention has uniquely recognized that neighboring transistors, even if turned off, tend to provide parasitic electrical characteristic values that distort the true electrical characteristic values of the functionally separate transistor being tested. These issues have been particularly observed in high-density arrays, such as those found in static random access memory (SRAM) devices.
The present invention has further recognized that the parasitic electrical characteristic values are particularly problematic when testing the leakage current of a single functionally separate transistor. Without being limited to any particular theory, the present invention theorizes that gate edge diode leakage (GEDL) from neighboring transistors (e.g., those transistors sharing a terminal, or drain in most instances, with the single functionally separate transistor being tested, as well as being located in the same active region) may be attributed with a significant portion of the parasitic leakage current.
Given the foregoing recognitions, the present invention acknowledged that the parasitic leakage current could be reduced, if not eliminated, by characterizing pairs of functionally separate transistors in a particular fashion. If the pair of functionally separate transistors are like transistors, the measured leakage current can be attributed equally to the two functionally separate transistors for determining the average leakage current for such transistors. If the pair of transistors are not like transistors, the leakage per transistor will be less precisely known. Nevertheless, the leakage per pair of transistors is still useful information for understanding the total leakage in a circuit.
Like transistors, as used herein, are transistors that have substantially identical properties, including without limitation, dopant types, dopant concentrations, material types, gate lengths, gate widths, work function, and other related properties. For example, in one situation wherein the transistors being tested form at least a portion of a SRAM device, the pair of like transistors could be a pair of like drive transistors, or in another example a pair of like pass transistors. Each of the pair of like transistors, whether they be like drive transistors or like pass transistors, are advantageously located in adjacent cells of the SRAM device. As adjacent cells of the SRAM device are typically mirror images of one another, the pair of like transistors are generally located proximate one another. It should be noted that other different types of like transistors could also be used.
Functionally separate, as used herein, means that at least one of the source, gate, drain or substrate of a transistor, when in an actual circuit that will be sold to a consumer, is connected to a different circuit node. Accordingly, functionally separate transistors do not include a multi-finger transistor structure operated in parallel to increase the drive current. However, functionally separate transistors might include a drive transistor located within a SRAM array, a pass gate transistor located within a SRAM array, a transistor in a NAND gate, a transistor in a multiplier array, etc. .
In an exemplary embodiment of the present invention, the pairs of functionally separate transistors are characterized by biasing a terminal between the pair of functionally separate transistors relative to gates of the pair and terminals outlying the pair. In this exemplary embodiment, the terminal between the pair is biased with respect to the substrate, where the bias may be non-zero, thus leakage is induced. However, the terminals outlying the pair are at the same bias with respect to the substrate, and thus little or no leakage is induced. Accordingly, the terminals outlying the pair act as an isolation region for the pair, and thereby reduce or eliminate the parasitic leakage current from adjacent transistors on opposing sides of the pair.
Using the aforementioned process, the actual leakage current associated with the pair may be measured. Thereafter, the actual measured leakage current of the pair may be divided by two to account for the two functionally separate transistors in the pair, thus obtaining an accurate measurement of the actual leakage current for each of the transistors. Because the pair of functionally separate transistors are like transistors, as defined above, the division of the actual measured leakage current of the pair is appropriate. For pairs of unlike functionally separate transistors, the above method, excluding the division by two, gives an accurate measurement of the total leakage of the pair. An accurate measure of the sum of leakage of two transistors is useful even without an allocation of the leakage to individual transistors.
Turning now to
In the exemplary embodiment shown, the shared terminal 135 is a source/drain region shared by the first and second transistors 120, 150. Similarly, the first outlying terminal 130 and the second outlying terminal 160 are source/drain regions of the first and second transistors 120, 150, respectively. As those skilled in the art appreciate, and as will be further discussed below with respect to
The electrical properties of the structure 100 may be tested by biasing the shared terminal 135 between the pair of like functionally separate transistors 110 relative to the first gate 125 and second gate 155 of the pair of like functionally separate transistors 110, as well as relative to the first outlying terminal 130 and second outlying terminal 160. For instance, by applying a nominal voltage to the shared terminal 135, while the first gate 125, second gate 155, first outlying terminal 130, second outlying terminal 160, and substrate 105 are held to ground, a leakage current associated with the pair of functionally separate transistors 110 can be obtained. In one exemplary embodiment, the voltage applied to the shared terminal 135 would range from about 0.5 volts to about 2.4 volts. However, other voltages outside of this range may nonetheless also be used.
In this exemplary embodiment, the shared terminal 135 between the pair 110 is biased with respect to the substrate 105, where the bias may be non-zero, thus diode leakage to the substrate is induced. However, the outlying terminals 130, 160 are at the same bias with respect to the substrate 105, and thus little or no diode leakage to the substrate is induced. Accordingly, the outlying terminals 130, 160 act as an isolation region for the pair 110, and thereby reduce or eliminate the parasitic leakage current from adjacent transistors on opposing sides of the pair 110. It should be noted that while
Turning now to
As those skilled in the art appreciate, the first transistor 220 includes a first gate 225, a source/drain region 230, and a shared source/drain region 235. Similarly, the second transistor 240 includes a second gate 245, the shared source/drain region 235 and a shared source/drain region 250. Likewise, the third transistor 260 includes a third gate 265, the shared source/drain region 250, and a shared source/drain region 270. Additionally, the fourth transistor 280 includes a fourth gate 285, the shared source/drain region 270 and a source/drain region 290. Each of the first, second, third and fourth transistors 220, 240, 260, 280 will also traditionally include channel implants, pocket/halo implants, etc., even though such features are not shown in the embodiment of
In the structure 200 of
It should be noted that while in this embodiment the pair of like functionally separate transistors 295 are drive transistors, the same theory would hold true if the pair of like functionally separate transistors 295 were pass transistors. For instance, if a third cell were located adjacent and below the second cell 215, the pair of like transistors between the second cell 215 and the third cell would be a pair of like pass transistors. The methodology of the present invention would work equally as well in this alternative embodiment. It should also be noted that other than the shared source/drain region 235 and the shared source/drain region 270, the remaining terminals of the first transistor 220 and the fourth transistor 280 remain open. In other words, no electrical connection is made to the remaining terminals of the first transistor 220 and the fourth transistor 280. Alternatively, terminal 230 and 290 could be held at the substrate potential. It should be noted that while
Turning now to
Using the test structure 300 of
Turning now to
Turning finally to
In an alternative embodiment, leakage may be measured for pairs comprising a drive transistor and a pass transistor, with the outer source/drain regions biased at the substrate potential. In this alternative embodiment, the measurement gives the combined leakage contribution of the driver and pass transistors without allocating the leakage to the individual transistors. In another alternative embodiment, leakage can be measured for four adjacent transistors comprising a first drive transistor, a first pass transistor, a second pass transistor, and a second drive transistor, where the outer source/drain regions of the two drive transistors are biased at the substrate potential. Those skilled in the art will recognize further similar embodiments.
Given the embodiment of the SRAM structure 500 illustrated in
Although the present invention has been described in detail, those skilled in the art should understand that they could make various changes, substitutions and alterations herein without departing from the spirit and scope of the invention in its broadest form.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4587481 *||Jun 16, 1983||May 6, 1986||Siemens Aktiengesellschaft||Arrangement for testing micro interconnections and a method for operating the same|
|US4835458 *||Nov 9, 1987||May 30, 1989||Intel Corporation||Signature analysis technique for defect characterization of CMOS static RAM cell failures|
|US5600578 *||Aug 2, 1993||Feb 4, 1997||Advanced Micro Devices, Inc.||Test method for predicting hot-carrier induced leakage over time in short-channel IGFETs and products designed in accordance with test results|
|US5686335 *||Jul 22, 1996||Nov 11, 1997||Taiwan Semiconductor Manufacturing Company, Ltd||Method of making high-performance and reliable thin film transistor (TFT) using plasma hydrogenation with a metal shield on the TFT channel|
|US5745405 *||Aug 26, 1996||Apr 28, 1998||Taiwan Semiconductor Manufacturing Company, Ltd||Process leakage evaluation and measurement method|
|US6175245 *||Jun 25, 1998||Jan 16, 2001||International Business Machines Corporation||CMOS SOI contact integrity test method|
|US6211693 *||Dec 15, 1998||Apr 3, 2001||Texas Instruments Incorporated||Testability circuit for cascode circuits used for high voltage interface|
|US6249138 *||Nov 23, 1999||Jun 19, 2001||United Microelectronics Corp.||Method for testing leakage current caused self-aligned silicide|
|US6388927 *||Feb 23, 2001||May 14, 2002||Cypress Semiconductor Corp.||Direct bit line-bit line defect detection test mode for SRAM|
|US6529421 *||Aug 28, 2001||Mar 4, 2003||Micron Technology, Inc.||SRAM array with temperature-compensated threshold voltage|
|US6552941 *||Jul 11, 2001||Apr 22, 2003||International Business Machines Corporation||Method and apparatus for identifying SRAM cells having weak pull-up PFETs|
|US6593590 *||Mar 28, 2002||Jul 15, 2003||Advanced Micro Devices, Inc.||Test structure apparatus for measuring standby current in flash memory devices|
|US6774656 *||Nov 1, 2001||Aug 10, 2004||International Business Machines Corporation||Self-test for leakage current of driver/receiver stages|
|US6930504 *||Jun 18, 2004||Aug 16, 2005||Renesas Technology Corp.||Semiconductor integrated circuit device|
|US6950359 *||Mar 28, 2003||Sep 27, 2005||Qualcomm Inc.||Memory bit line leakage repair|
|US7149114 *||Mar 17, 2004||Dec 12, 2006||Cypress Semiconductor Corp.||Latch circuit and method for writing and reading volatile and non-volatile data to and from the latch|
|US7155359 *||Jul 2, 2004||Dec 26, 2006||Advanced Micro Devices, Inc.||Determination of device failure characteristic|
|U.S. Classification||365/201, 365/200, 365/189.08|
|Cooperative Classification||G11C2029/5006, G11C29/50, G11C11/417, G01R31/3008, G11C11/41|
|European Classification||G11C11/417, G11C29/50, G01R31/30C1|
|Nov 8, 2005||AS||Assignment|
Owner name: TEXAS INSTRUMENTS INC., TEXAS
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:HOUSTON, THEODORE W.;DENG, XIAOWEI;GELSOMINI, TITO;REEL/FRAME:017208/0645
Effective date: 20051101
|Aug 28, 2012||FPAY||Fee payment|
Year of fee payment: 4